Not Just Lumber—Using Wood in the Sustainable Future of Materials, Chemicals, and Fuels

Home , Composite lumber

JOM, Vol. 68, No. 9, 2016 DOI: 10.1007/s11837-016-2026-7 Ó 2016 The Minerals, Metals & Materials Society (outside the U.S.)

JOSEPH E. JAKES,1,8 XAVIER ARZOLA,1,2 RICK BERGMAN,3 PETER CIESIELSKI,4 CHRISTOPHER G. HUNT,1 NIMA RAHBAR,5 MANDLA TSHABALALA,1 ALEX C. WIEDENHOEFT,6 and SAMUEL L. ZELINKA7

1.—Forest Biopolymers Science and Engineering, USDA Forest Service, Forest Products Laboratory, One Gifford Pinchot Drive, Madison, WI 53726, USA. 2.—Materials Science and Engineering, Univ- ersity of Wisconsin-Madison, 1509 University Avenue, Madison, WI 53706, USA. 3.—Economics, Statistics and Life Cycle Analysis Research, USDA Forest Service, Forest Products Laboratory, One Gifford Pinchot Drive, Madison, WI 53726, USA. 4.—Biosciences Center, National Renewable Energy Laboratory, 15013 Denver W Parkway, Golden, CO 80401, USA. 5.—Department of Civil and Envi- ronmental Engineering, Worcester Polytechnic Institute, 100 Institute Road, Worcester, MA 01609, USA. 6.—Center for Wood Anatomy Research, USDA Forest Service, Forest Products Laboratory, One Gifford Pinchot Drive, Madison, WI 53726, USA. 7.—Building and Fire Sciences, USDA Forest Ser- vice, Forest Products Laboratory, One Gifford Pinchot Drive, Madison, WI 53726, USA. 8.—e-mail: [email protected]

Forest-derived biomaterials can play an integral role in a sustainable and renewable future. Research across a range of disciplines is required to develop the knowledge necessary to overcome the challenges of incorporating more renewable forest resources in materials, chemicals, and fuels. We focus on wood speciﬁcally because in our view, better characterization of wood as a raw material and as a feedstock will lead to its increased utilization. We ﬁrst give an overview of wood structure and chemical composition and then highlight current topics in forest products research, including (1) industrial chemicals, biofuels, and energy from woody materials; (2) wood-based activated carbon and carbon nanostructures; (3) development of improved wood protection treatments; (4) massive timber construction; (5) wood as a bioinspiring material; and (6) atomic simulations of wood polymers. We conclude with a discussion of the sustainability of wood as a renewable forest resource.

play an increasing role in the sustainability of our WOOD’S 390 MILLIONTH BIRTHDAY future materials through both expansion of uses for Wood is one of the major innovations of land current forest products and development of alter- plants ‘‘invented’’ some 390 million years ago. Wood natives to materials currently derived from nonre- enabled individual plants to increase their stature newable resources. Life cycle assessments, which and persistence in the environment, facilitating the categorize energy consumption and emission pro- ability of trees to play roles in landscape change and ﬁles for products over their whole life cycle,1,2 biome composition and to fundamentally alter the consistently show that many wood-based materials global cycling of water, minerals, and carbon. Since use less fossil fuels to produce than do competing prehistoric times, humans have found wood invalu- materials.3,4 Using wood products can also lower able for meeting many of their needs, including atmospheric carbon dioxide levels because growing energy, tools, and shelter—not surprising given its forests capture carbon and harvested wood products near-ubiquity, versatility, and renewability. Only in store the accumulated carbon while in service. the comparatively recent past have other nonre- Historically, wood research has been the purview newable natural resources, such as fossil fuels or of wood technologists; however, we see the future of metal ores, become the resources of choice to meet wood research as interdisciplinary, collaborative, our material needs. We postulate that wood will quantitative, and meaningful, both scientiﬁcally

(Published online July 21, 2016) 2395 2396 Jakes, Arzola, Bergman, Ciesielski, Hunt, Rahbar, Tshabalala, Wiedenhoeft, and Zelinka and societally. In this review, we brieﬂy introduce wood as a material, present current trends in forest products research advancing and expanding the role of wood-based materials in our sustainable materials future, and further discuss the sustainability of forests for wood resources. One notable exception from this review is cellulose nanomaterials research. Nevertheless, cellulose nanomaterials is the subject of another review in this JOM special edition.5

WOOD STRUCTURE AND COMPOSITION Wood is an anisotropic cellular tissue, the structure and properties of which are derived to solve plant problems—sap conduction, mechanical support, and storage and synthesis of biochemicals—in space and over time.6,7 The properties of wood that make it desirable as a natural resource, such as its exceptional strength-to-weight ratio or the richness, breadth, and subtlety of cell wall biopolymers, are thus predicated on the biological origins and functions of wood in the living plant. Wood solves the simultaneous need for conduction, support, and storage by combining arrays of cells of different sizes, shapes, chemistries, and functions, and it does so using large numbers of cells integrated over long distances (from root to branch). The bulk of wood is made of cells of an axial system oriented parallel to the long axis of the major tree components, including roots, trunks, and branches. The axial system is primarily implicated in conduction and mechanical support in most woods, with a lesser role in storage and synthesis. The other cell system in wood is the radial system, which is composed of cells oriented perpendicular to both the long axis of the tree component and growth rings. The radial system is primarily implicated in (connected to) storage and synthesis. The two basic body plans for wood are that of softwoods (such as pines, cedars, and spruces; Fig. 1a) and that of hardwoods (such as elms, oaks, ashes, maples, and birches; Fig. 1b). Both softwoods and hardwoods share a fundamental anatomy based on relative relationships among the axial system, the radial system, and the typically round cross sections of roots, trunks, and branches. This topol- ogy gives rise to three distinct planes of view in wood: transverse, radial, and tangential (Fig. 1b). The transverse plane is the cross section that is perpendicular to the axial system. Radial and tangential planes are cut parallel to the axial system and oriented relative to the cells of radial system, parallel to the rays for a radial-longitudinal plane and perpendicular to the rays for a tangential- Fig. 1. From tree to cell wall layers. Stylized tree silhouettes for a longitudinal plane. generic softwood (a) and a generic hardwood (b). Variations in cell The primary differences between softwoods and types, size, and distribution within the axial system gives rise to growth hardwoods are found in the cells of the axial system rings, which are used to define the three primary planes of wood (Fig. 1d and e, respectively)—the radial systems of structure (c). Photomicrographs of transverse sections of a typical softwood (d) and a typical hardwood (e). The idealized lamellar structure each are, although different in their specific anat- of cells with secondary walls (f) from the lumen outward, secondary cell omy, generally similar in cell type, function, and wall layers (S3, S2, and S1), primary cell wall, and middle lamella. Not Just Lumber—Using Wood in the Sustainable Future of Materials, Chemicals, and Fuels 2397 distribution. Cells in wood, especially in dry wood, have two domains: the cell wall (the physical substance of the cell) and the cell lumen (interior void volume or air space delimited by the cell wall). Softwoods are characterized by a comparatively simple wood structure—greater than 90% of softwoods by volume are typically a single type of cell, the tracheid. Tracheids are long, thin cells serving double duty of conduction and mechanical support. Although the softwood-type body plan first evolved more than 390 million years ago, it continues to be an efficient, competitive design, boasting some of the tallest (Sequoia sempervirens, greater than 115 m tall) and oldest (Pinus longaeva, more than 4000 years old) living trees in the world. Hardwoods Fig. 2. Examples of the molecular precursors of wood polymers are characterized by a cellular-level division of labor including (a) six-carbon (6C) sugar D-glucose (cellulose), (b) 6C between dedicated conductive cells (vessel ele- sugar D-mannose (hemicellulose), (c) five-carbon (5C) sugar D-xy- ments) and mechanical cells (fibers). A third cell lose (hemicellulose), and (d) Sinapyl alcohol (lignin). type, specialized for storage and synthesis, is also found in most hardwoods, axial parenchyma. The total volume of each cell type in hardwoods varies the most abundant biopolymer on Earth, is a linear widely from species to species, and because hard- polysaccharide made from the simple sugar D-glu- wood species are found in most habitats that include cose (Fig. 2a). Individual cellulose chains are held woody plants, this diversity of patterns is the basis together by intermolecular and intramolecular of the wide variety of wood structure and properties hydrogen bonds to form the semicrystalline cellu- found in nature. lose elementary fibrils. The structure of cellulose Most woody cells have secondary cell walls. microfibrils in any cell wall layer is proposed to Proceeding from the lumen of the cell outward, the consist of a 3 by 4 array of approximately 3-nm- first layer encountered is the innermost layer of the diameter semicrystalline cellulose elementary fib- secondary cell wall, the thin S3. This is followed by rils with thin layers of less-ordered structures integrated into the cellulose microfibrils between the thick S2, the thin S1, the primary cell wall, and 10,11 then the middle lamella. Secondary cell walls are the elementary fibrils. Hemicelluloses are amor- nano-fiber-reinforced composites of highly oriented phous, highly branched polymers composed primar- semicrystalline cellulose microfibrils embedded in a ily of the sugars D-mannose (Fig. 2b), D-galactose, D- matrix of amorphous cellulose, hemicelluloses, and xylose (Fig. 2c), L-arabinose, and D-glucuronic acid. lignin. Elucidating the fine details of the cell wall Lignin, which is the second most abundant biopoly- nanostructure is still an open area of research.8,9 mer on Earth, is an amorphous, highly cross-linked The S2 layer is much thicker than the S1 and S3 aromatic polymer polymerized through free radical layers and therefore has a larger influence on wood reactions from sinapyl alcohol (Fig. 2d), coniferyl properties than have the other cell wall layers. The alcohol, and p-cumaryl alcohol. Carbon–carbon (C– helical angle the cellulose microfibrils make with C) and carbon–oxygen–carbon (C–O–C; ether) bonds the longitudinal cell axis is called the microfibril are formed during lignin polymerization. The over- angle (Fig. 1f) and in the S3 layer is typically >70°, all chemical composition in wood can vary between in the S2 is typically low (5°–30°), and is typically tree species and type of wood tissue within a given 50°–70° in the S1.6 The primary wall is very thin tree. and characterized by a randomly oriented web of Water is another important component of wood cellulose microfibrils.10,11 The middle lamella fills in that affects numerous wood properties, including mechanical, thermal, electrical, dimensional stabil- the regions between cells (Fig. 1f) and consists of 15 about 20% hemicelluloses embedded as an irregu- ity, and durability. For example, changes in the lar, interconnecting network in a matrix of lig- amount of water in wood can cause changes of nin.6,12 In most cases, the primary cell wall, middle nearly 10 orders of magnitude in electrical proper- lamella, and primary cell wall of the adjacent cell ties and anisotropic dimensional changes in bulk wall are not separable as distinct layers; these three wood of approximately 10, 5, and 0.1% in the layers together are known as the compound middle tangential, radial, and longitudinal directions, lamella. In typical wood cells, cellulose microfibrils respectively. The accessible hydroxyl (–OH) and are about 15–20 nm in diameter.10,13 other polar chemical groups (such as ether and Chemically, on a dry basis, secondary wood cell carbonyl linkages) in the amorphous cellulose, walls consist of about 35% highly oriented, hemicelluloses, and lignin readily adsorb water. semicrystalline cellulose microfibrils embedded in The moisture content (MC) of wood is defined as a matrix of amorphous cellulose (20%), hemicellu- water mass divided by oven-dried wood mass and loses (30%), and lignin (15%).14 Cellulose, which is when not in contact with liquid water depends on 2398 Jakes, Arzola, Bergman, Ciesielski, Hunt, Rahbar, Tshabalala, Wiedenhoeft, and Zelinka ambient temperature and relative humidity (RH). aromatic compounds; and char. For liquid fuel Below the fiber saturation point, approximately 30% production, the desired product is obtained by to 40% MC, all water in wood is bound by inter- condensation of the heavy gasses into a liquid called molecular attractions within the wood cell wall ‘‘pyrolysis oil’’ or ‘‘bio-oil.’’ The competing reactions polymers. At higher MCs, free water forms in wood that occur during pyrolysis govern the formation of cavities, such as cell lumina and open cell corners. light gasses, condensable vapors, and char. These reactions occur at different rates and with different CURRENT TOPICS activation energies and thereby provide an opportunity to tune the product distribution by control- Industrial Chemicals, Biofuels, and Energy ling process conditions. For example, production of from Woody Materials condensable vapors occurs with a lower activation Energy-rich biopolymers that constitute woody energy and by more rapid kinetics relative to char cell walls present an important opportunity for formation; therefore, very rapid heating in tandem production of renewable fuels and chemicals. with short residence times in the reactor shift Strategies for production of liquid fuels and chem- product yields toward condensable gases while icals from woody feedstocks can be generally divided minimizing char formation.23 Raw bio-oil from into two classes consisting of biochemical pathways pyrolysis has high oxygen content and therefore and thermochemical pathways. Biochemical path- relatively low energy content compared with petro- ways typically aim to depolymerize cellulose and leum-derived liquid fuels. Efforts to address this hemicelluloses into C6 and C5 sugars (for example, challenge have resulted in processes such as cat- Fig. 2a–c) using cocktails of hydrolytic enzymes. alytic fast pyrolysis,24 where raw pyrolysis vapors The process of hydrolyzing polysaccharides to sol- are passed through a deoxygenation catalyst prior uble sugars is called ‘‘saccharification.’’ Challenges to condensation, and fast hydropyrolysis where associated with enzymatic deconstruction of bio- pyrolysis is performed under a significant overpres- mass are contributed by the complex lignocellulosic sure of hydrogen gas, which removes oxygen from 16 25 structure of wood cell walls. In lignocellulosic the biomass primarily in the form of H2O. Fast biofuels research, resistance to saccharification is pyrolysis also serves as the first step in gasification, often termed ‘‘recalcitrance.’’ Treatments that another thermochemical conversion strategy that expose the biomass to elevated temperatures and produces syngas (i.e., a mixture of H2, CO, and CO2) chemicals such as dilute sulfuric acid,17 mixtures of from biomass by controlled reaction of the gaseous organic solvents (such as methyl isobutyl ketone, products with steam and/or oxygen at a high ethanol, and water), which is also called organo- temperature (700°C to 1000°C). Although gasifica- solv,18 and sulfite19 have been shown to overcome tion plants for both coal and biomass have existed recalcitrance and enhance saccharification yields of for several decades,26 this technology has experi- the resulting residues by increasing the accessibility enced a resurgence of attention from the biofuels of cellulose to enzymes. The solubilized sugar community due in part to recent projections that stream produced from enzymatic hydrolysis, called gasification technology could contribute signifi- hydrolysate, is then subjected to additional biolog- cantly to future renewable heat and power ical or chemical catalytic upgrading to produce generation.27 desired fuel molecules. The traditional process used Although the importance of producing renewable for this purpose is biological fermentation to pro- fuels from woody feedstocks is widely recognized, duce ethanol.20 More recent efforts aim to produce recent techno-economic evaluations of biorefinery fuel molecules other than ethanol from hydrolysate processes have highlighted the importance of pro- by strategies wherein sugar-derived compounds ducing additional valuable chemical co-products such as furans, organic acids, and polyols serve as from wood polymers in order for biorefineries to intermediates for the production of hydrocarbons achieve economic self-sustainability. For example, through reduction and carbon–carbon coupling in the context of a biochemical biorefinery, the reactions.21,22 importance of lignin valorization has been specifi- Thermochemical conversion pathways differ from cally identified.28 Conventional biorefinery models biochemical conversion pathways in that they rely combust lignin to generate process heat; however, primarily on elevated temperatures, rather than on recent advances in biochemical processing of lignin biocatalysts, to deconstruct biomass into small have resulted in genetically engineered microbes molecules. Technologies such as fast pyrolysis, capable of metabolizing lignin into valuable plat- gasification, and hydrothermal liquefaction fall form chemicals.29 Alternative approaches have under this category. Fast pyrolysis is a process developed one-pot catalytic systems for depolymer- wherein biomass is rapidly heated to temperatures ization and subsequent conversion of lignin from of 400°C to 600°C in an inert environment where poplar wood into high-value specialty chemicals, the feedstock decomposes into light gases including such as methoxypropylphenols, with high yield and 30 CO, CO2, and small hydrocarbons; condensable selectivity. Such strategies are prompting gasses comprising heavier hydrocarbons and researchers to re-envision the modern biorefinery Not Just Lumber—Using Wood in the Sustainable Future of Materials, Chemicals, and Fuels 2399 to a process where biomass fractionation and con- networks can be produced by a process involving the version of lignin to valuable chemical co-products rapid freezing of an aqueous lignin solution, fol- play a central role.31 lowed by sublimation of the resultant ice, to form a uniform network made up of individual intercon- Wood-Based Activated Carbon and Carbon nected lignin nanofibers. Carbonization of lignin Nanostructures nanofibers yields a similarly structured CNF network.36 Researchers are also investigating first The earliest known use of wood-based carbon spinning lignin-based fibers followed by carboniza- materials is 3750 BC when the ancient Egyptians tion to create single CNFs.37 Lignin-based CNFs and Sumerians used wood chars for the reduction of seem to be promising and may play a role in the copper, zinc, and tin ores in the manufacture of lignin valorization needed to achieve economic self- bronze and as domestic smokeless fuel.32 Since sustainability in biorefineries. then, many other forms of wood-based carbon CNTs are a particular form of fullerene, first materials have been developed, including activated reported by Iijima.38 They are tubular structures carbons and carbon nanostructures. that can be 1–2 nm in diameter and ‡1mmin Activated carbons are used in numerous commer- length. CNTs have great tensile strength and are cial applications including water treatment, CO 2 considered to be 100 times stronger than steel, while capture, energy storage, supercapacitors, and being only one sixth of its weight, making them heterogeneous catalysis.33 They are also used in potentially the strongest, smallest fiber known. the food industry for purification of oils and fats and They also exhibit high electrical conductivity, high alcohol drinks, as well as in the chemical and surface area, unique electronic properties, and pharmaceutical industries for gas and drug purifi- potentially high molecular adsorption capacity.39 cation. The basic procedure to produce activated Applications currently being investigated include carbon is to first carbonize the material at high polymer composites (conductive and structural temperatures under inert environments to produce filler), electromagnetic shielding, electron field emit- a microporous carbonaceous mass with a high ters (flat panel displays), supercapacitors, batteries, surface-to-volume ratio. The surfaces in the micro- hydrogen storage, and structural composites. Inter- porous carbon can then be further activated by estingly, CNTs have been produced from wood fiber processes, such as steam activation, to increase the using a low-temperature process, which included amount and specificity of its absorptivity. Although continuous oxidization at 240°C and cyclic oxidation activated carbons can be produced from almost any at 400°C. The inside diameter of the CNTs was carbon-based raw material, it is more cost effective approximately 4–5 nm, and the outside diameter and environmentally desirable to produce them ranged from 10 nm to 20 nm. No CNTs were from sustainable or waste materials. Activated produced when pure lignin or pure cellulose were carbons produced from wood or similar lignocellu- tested, indicating that the cell wall nanostructure, losic materials show great promise because they can likely the cellulose elementary fibrils, plays an possess highly developed microporous structures important role in CNT formation. Apparently, the that are likely enhanced by naturally occurring differential ablation properties of the major cell wall nanostructures in the cell walls. For example, components, cellulose, hemicellulose, and lignin, are activated carbons produced from coconut shells critical for the formation of CNTs at comparatively have long been shown to have high volumes of low temperatures.40 micropores, making them a commonly used raw material for applications where high adsorption Massive Timber Construction (Tall Wood capacity is needed, such as water filters. Neverthe- Building) less, the natural variability in precursors like wood poses challenges because it can result in production Wood has been used as a building material for of activated carbons without the necessary control millennia. In the United States, most single-family over important properties, which include porosity, homes are built using ‘‘stick frame’’ wood construc- morphology, mechanical properties, and surface tion with wooden studs carrying the vertical loads chemistry.34 Fortunately, development of more flex- and wood panel products, such as plywood or ible and robust routes to new carbon materials orientated strandboard (OSB), attached to the out- derived from renewable resources like wood is side for lateral and transverse loads. Interest has becoming a topic of increased interest and improve- recently increased in building nonresidential and ments are expected.33 multistory residential wood buildings using larger Carbon nanofibers (CNFs) and carbon nanotubes timber components. For example, in 2008, a nine- (CNTs) are examples of carbon nanostructures that story high-rise was completed in London using wood can be derived from wood. Based on the fact that structural members. The structure, called ‘‘24 Mur- isolated lignin has the largest carbon content among ray Grove,’’ became the world’s tallest residential wood components, CNFs have been produced from wood structure.41 Timber buildings from 9 to 20 lignins and used in fabrication of electrodes for stories high are also planned in New York, Port- lithium-ion batteries and supercapacitors.35 CNF land, Canada, Norway, and Austria.42 2400 Jakes, Arzola, Bergman, Ciesielski, Hunt, Rahbar, Tshabalala, Wiedenhoeft, and Zelinka

structural composite lumber and CLT are inher- ently greater than that of solid wood as a result of the different orientations of individual wood components within the larger timber.44 The dimensional stability of these composites is extremely important in high-rise construction where small, moisture-induced deformations are additive through the height of the building. Finally, new fastening systems such as self-tapping screws and epoxied steel rods have been developed to connect mass timber composites. These technologies have improved the moment-resisting capacity and seis- mic performance of timber joints.45–47

Developing Improved Wood Protection Treat- Fig. 3. Massive timber composites. From left to right: cross-laminated timber (CLT), structural composite lumber (LVL), and glulam. ments Wood is widely used in construction in North America, and replacing decayed or moldy wood costs Tall wood buildings can be realized through the billions of dollars per year.48 In addition, concerns use of mass timber construction. ‘‘Mass timber’’ is a about decay sometimes prevent wood from being class of wood composites that includes glue-lami- used. Clearly, improved decay resistance without nated timber (glulam), structural composite lumber, significant added cost or detrimental environmental and cross-laminated timber (CLT) (Fig. 3). Glulam effects would increase wood utilization. consists of stacked lumber glued along the grain to In wood construction, the primary agents of wood form a massive beam. Structural composite lumber decay are brown rot fungi, which are adapted to this is a general term that can refer to wood composites recalcitrant food source and produce a severe loss of made from veneers or flakes of wood aligned so that wood strength during early decay. Whereas fungi the grain is parallel to form a billet. Structural and their secreted enzymes fit easily into the lumina composite lumber made from veneers is referred to of wood cells (Fig. 1f), the pores in the nanostruc- as laminated veneer lumber (LVL). Structural com- ture of wood cell walls are too small for even the posite lumber made from flakes is referred to as smallest enzymes to gain access. Brown rot fungi parallel strand lumber (PSL), laminated strand are able to attack wood by producing low-molecular- lumber (LSL), or orientated strand lumber (OSL), weight oxidants or oxidant precursors, much smal- depending on the size and aspect ratio of the flakes ler than enzymes, which diffuse through the wood used to make the composite. CLT has attracted the cell wall, to oxidize and cleave cell wall polymers. most attention for its use in tall wood buildings. This releases soluble sugars that can be taken up as CLT is a massive panel product made from layers of a carbon source by the fungus.49 dimensional lumber where the long axis of each If wood is wet enough to support microbial layer is orientated 90° from the previous layer. CLT activity for a prolonged period, or cyclically wet panels may vary from approximately 50 mm to and dry often enough, decay will occur unless the 500 mm in thickness, can be manufactured up to wood is treated with wood preservatives or chemi- 18 m long, and are typically delivered to the job site cally modified because the spores of wood decay with all openings for windows and doors precut.43 organisms are ubiquitous. Wood preservatives are CLT is the main component of the 24 Murray Grove chemicals added to the wood to protect it from fungi building and figures prominently in the design of and insect attack and frequently contain copper.50 the winners of the United States Tall Wood Building Wood preservatives are registered pesticides, and Prize competition. future regulation and their availability depend on Mass timber composites were developed through- maintaining their registration with environmental out the 20th century; however, recent advances in regulators, such as the Environmental Protection manufacturing and fastening systems have made Agency in the United States.51 An alternative to them easier to use and more attractive as a building wood preservatives is a chemical or thermal modi- material. The rise of large computer numerically fication of wood polymers to alter the wood chemical controlled (CNC) routers now allows for massive or physical structure to inhibit decay. Such modified CLT panels to be prefabricated to a high precision at wood products have significant market penetration the mill before being delivered to the jobsite. in Europe and are commercially available world- Massive timber composites are also being made wide. Commercialized techniques include thermal under tighter control of RH, which is important modification, impregnation and in situ polymeriza- because wood changes dimension with changes in tion of furfuryl alcohol, and acetylation by impreg- MC.15 Furthermore, the dimensional stability of nation and reaction with acetic anhydride. Not Just Lumber—Using Wood in the Sustainable Future of Materials, Chemicals, and Fuels 2401

Current questions include how chemical modifi- Also, multifunctional artificial materials with tun- cations inhibit decay and how lower cost-effective able mechanical performances and a propensity for modification procedures can be developed. The self-healing have been created from nanofibrillated mechanism of decay prevention in acetylated wood cellulose and cationic poly(vinyl amine) by mimick- has been the most extensively studied of all chem- ing the cell wall nanostructure.61 ical modifications, but it is still debated.52 Acetyla- Moisture responses in wood are also being studied tion becomes effective at approximately 10% to 20% with the motivation to inspire new stimuli-respon- weight gain by replacing wood polymer hydroxyl (– sive and multifunctional materials. Stimuli-respon- OH) groups (see Fig. 2) with acetoxy (–OCOCH3) sive materials change according to the environment groups.53 Wood polymers swell with moisture, they are in and are designed to be sensitive to a increasing their free volume. In acetylated wood, variety of stimuli, such as temperature, heat, sol- the new acetoxy groups occupy some of this volume vent, light, moisture, and electric or magnetic fields. and make the wood polymers less hydrophilic, Such materials have applications in areas as diverse thereby lowering the amount of water that can be as energy harvesting, sensors, drug delivery sys- absorbed in wood cell walls. Among other theories, tems, biomimetic robotics, and artificial mus- decay inhibition has been attributed to this lowered cles.62–66 In trees, researchers have studied cell wall MC, which may result in lower diffusion moisture-activated unidirectional movements, such rates for fungal oxidants through wood cell walls.52 as pine cone opening67 and bending of tree Nevertheless, the connection between MC and branches.68 Models have been developed relating diffusion inside the cell wall remains unclear. these unidirectional movements to moisture-in- It was recently proposed that ion diffusion inside duced swelling in the amorphous components of wood is a percolation-controlled phenomenon,54 and wood cell walls and the cellular organization of wood recent experimental results showed a MC threshold tissue with different cellulose microfibril angles. below which ions did not diffuse through wood cell Additionally, wood slivers consisting of a few soft- walls.55,56 This led to the hypothesis that the key to wood tracheids were found to be moisture-activated stopping fungal oxidant diffusion lies in preventing torsional actuators that can reversibly twist multi- the formation of a percolating network of diffusion ple revolutions per centimeter length and produce domains within the cell wall.57 This hypothesis, specific torque higher than that of an electric although unproven, suggests that it might be pos- motor.55,69 The twist is caused by swelling between sible to develop new, low-cost modifications that the helically wound cellulose microfibrils in the S2 inhibit decay by modifying only those cell wall secondary cell wall (Fig. 1f). The wood slivers also polymers implicated in cell wall diffusion. possess moisture-activated shape memory twist capabilities.55,69 Individual wood cell walls were Wood as a Bioinspiring Material also discovered to be moisture-activated materials for chemical transport. Ions implanted into cell In a living tree, wood is a multifunctional mate- walls were observed to diffuse only above a thresh- rial simultaneously fulfilling the tree’s needs for sap old MC.55,56 Both the shape memory and chemical conduction, mechanical support, and storage and transport thresholds are likely controlled by the synthesis of biochemicals. As with many of Nature’s moisture-dependent glass transition of hemicellu- materials, the performance of wood in a living tree loses in the 60% to 80% RH range at room temper- is unrivaled by current synthetic materials and the ature.70–72 Glass transitions are often responsible study of Nature’s design secrets continues to be a for shape memory and stimuli-responsive behavior reliable source to improve or develop new technolo- in polymers. When considered in bulk as a polymer, gies. Researchers continue striving to understand wood does not experience the macro-scale mechan- the structure and properties of wood for inspiration ical softening at the glass transition like other to improve or create new materials, structural polymers, which is expected, given that wood must designs, and theoretical formulations. perform its mechanical functions above the MC- As a complex composite, wood is providing aca- inducing glass transition in the living tree. An demic inspiration for development of modeling and improved understanding of the wood cell wall computational frameworks to solve optimization nanostructure may lead to biomimetic polymer problems related to mechanical properties of com- smart materials with improved mechanical plex heterogeneous materials.58 Branch-trunk properties. joints in trees are complex orthotropic, fiber-reinforced composites with extraordinary strength and Applications of Atomistic Simulations in toughness. Biomimetic joints are being developed as Understanding Small-Scale Material Proper- an alternative approach to solve T-joint problems ties of Wood for potential use in lightweight aircraft structures.59 Renewable packaging films inspired by the Wood is a hierarchal material, and understanding nanostructure of wood cell walls have been formed its structure and properties requires thorough by mixing cross-linked galactoglucomannan–lignin knowledge of material properties down to the nano- polymers, microfibrillated cellulose, and glycerol.60 and molecular-scales. Many properties at these 2402 Jakes, Arzola, Bergman, Ciesielski, Hunt, Rahbar, Tshabalala, Wiedenhoeft, and Zelinka length scales cannot be assessed experimentally. Overall, as the preceding examples of studies Fortunately, numerical methods, such as atomistic clearly show, atomistic simulations provide a plat- simulations, are emerging as powerful tools at these form for studying structures, interactions, and small scales for simulating underlying physical and properties at the nano- and molecular scales. The chemical mechanisms or predicting the thermody- role of simulations is expected to continue to namic properties of wood cell walls from the increase as research increases. Nevertheless, atomistic level. This motivates the use of molecular because wood is such a complex and unknown dynamics (MD) as it enables researchers to build material at these small scales, a key to increasing bottom-up models starting from the chemical struc- the potential benefits of simulations in forest prod- ture of the basic constituents of wood. ucts research will be developing appropriate exper- There has been much focus on understanding the iments to further validate simulated results. structure and properties of crystalline nanocellulose because it is the primary nanoscale component SUSTAINABLE FOREST MANAGEMENT FOR responsible for the superb mechanical properties of RENEWABLE WOOD RESOURCES wood. For example, united-atom MD simulations The sustainable management of forests is com- were used to quantify changes in different mor- plex because it must consider not only the regener- phologies of cellulose to reveal that the semicrys- ation of trees harvested for material usage but also talline phase may be an intermediate, kinetically the maintenance of benefits that are both societal arrested phase formed at amorphous cellulose for- (e.g., recreation, food, and spiritual) and ecological mation.73 MD simulation showed in Ib crystalline (e.g., climate regulation, biodiversity, and protec- cellulose that elastic modulus, Poisson’s ratio, yield tion of soil and water resources). Based on analyses stress and strain, and ultimate stress and strain are of the Global Forest Resources Assessment 2015 of highly anisotropic, and that although properties the Food and Agriculture Organization of the that describe elastic behavior of the material are United Nations, forests reportedly cover 31% of independent of strain rate, yield and ultimate the today’s world land surfaces78 and efforts to properties increase with increasing strain rate.74 sustainably manage forests, including forest man- In another MD study,75 the fracture energy of agement certification programs, continue to crystalline cellulose was found to depend on crystal increase worldwide.79 Certification programs came width, due to edge defects that significantly reduce about to demonstrate the commercial viability of a the fracture energy of small crystals but have a ‘‘sustainable’’ alternative to historical practices that negligible effect beyond a critical width. Remark- viewed forests as an infinite resource. Currently, ably, ideal dimensions optimizing fracture energy about 10% of the world’s forests are in a certification are found to be similar to the common dimensions of program,79 and increased international awareness crystalline nanocellulose found in nature, suggest- of and demand for chain of custody and supply chain ing a natural optimization of structure. verification in wood and wood products80 further Other researchers have focused on the assembly supports a transition toward sustainable forest and interactions of wood polymers in the wood products production. With enhanced and interdisci- nanostructure because the overall material proper- plinary research into wood and wood products, we ties of wood are a direct function of these currently have the expectation that high-value products and unknown nanostructures and interactions. Some of feedstocks from forest materials will help increase the earliest MD simulations studied interactions of the monetary value of forests and, thus, support lignin models with the surface of cellulose microfib- their sustainable management. rils. Attractions between lignin and cellulose microfibril surfaces that were observed suggested CONCLUSION that the polysaccharide components of the cell wall may be influencing deposition of lignin.76 In another The advancement of wood science is critical for study, lignin adsorption was found to likely be the development of sustainable materials, fuels, and dependent on crystal orientation of the microfibril.77 chemicals from forest resources. It is also invaluable In a more recent study, bamboo molecular models of to researchers who draw inspiration from the lignin, hemicellulose, and lignin carbohydrate com- elegant, hierarchical structure of wood for develop- plex (LCC) structures were used to study elastic ing advanced materials and structures. Numerous moduli and adhesion energies between these mate- synergies exist between different research areas rials and cellulose microfibril faces under dry con- that should be exploited through multidisciplinary ditions. It was shown that the hemicellulose model teams to accelerate efforts. For example, biorefinery has stronger mechanical properties than lignin, researchers would benefit from improved biomi- whereas lignin exhibits greater tendency to adhere metic deconstruction pathways if the mechanisms to cellulose microfibrils. The study suggested that used by decay fungi to deconstruct the recalcitrant the abundance of hydrogen bonds in hemicellulose wood cell walls into usable energy were better chains is responsible for improving the mechanical understood. Similarly, assuming controlling chem- behavior of LCC. ical transport through the cell wall is the key to Not Just Lumber—Using Wood in the Sustainable Future of Materials, Chemicals, and Fuels 2403 preventing the onset of wood decay, a collaboration 18. J.J. Bozell, S.K. Black, M. Myers, D. Cahill, W.P. Miller, between researchers developing wood protection and S. Park, Biomass Bioenergy 35, 4197 (2011). 19. J.Y. Zhu, X.J. Pan, G.S. Wang, and R. Gleisner, Bioresour. treatments and multifunctional smart chemical Technol. 100, 2411 (2009). transport membranes would also be mutually ben- 20. R. Gupta, K.K. Sharma, and R.C. Kuhad, Bioresour. eficial. Such collaborations between seemingly dif- Technol. 100, 1214 (2009). ferent research areas are key to accelerating wood 21. J.C. Serrano-Ruiz and J.A. Dumesic, Energy Environ. Sci. science and positioning it in the future with ample 4, 83 (2011). 22. J.J. Bozell and G.R. Petersen, Green Chem. 12, 539 (2010). opportunity for innovative, meaningful contribu- 23. A.V. Bridgwater, J. Anal. Appl. Pyrolysis 51, 3 (1999). tions across a broad spectrum of material science 24. C.C. Mukarakate, X. Zhang, A.R. Stanton, D.J. Robichaud, research. P.N. Ciesielski, K. Malhotra, B.S. Donohoe, E. Gjersing, R.J. Evans, D.S. Heroux, R. Richards, K. Iisa, and M.R. ACKNOWLEDGEMENTS Nimlos, Green Chem. 16, 1444 (2014). 25. V.K. Venkatakrishnan, J.C. Degenstein, A.D. Smeltz, W.N. J.E.J. acknowledges funding from 2011 USDA Delgass, R. Agrawal, and F.H. Ribeiro, Green Chem. 16, PECASE awards. Financial support for P.N.C. was 792 (2014). provided by the Computational Pyrolysis Consor- 26. A.J. Minchener, Fuel 84, 2222 (2005). tium (CPC) funded by the Bioenergy Technologies 27. J. Ahrenfeldt, T.P. Thomsen, U. Henriksen, and L.R. Clausen, Appl. Therm. Eng. 50, 1407 (2013). Office (BETO) of the U.S. Department of Energy. 28. A.J. Ragauskas, G.T. Beckham, M.J. Biddy, R. Chandra, F. We also thank Steve Schmeiding from the USDA Chen, M.F. Davis, B.H. Davison, R.A. Dixon, P. Gilna, M. Forest Products Laboratory for the photo used in Keller, P. Langan, A.K. Naskar, J.N. Saddler, T.J. Fig. 3. Tschaplinski, G.A. Tuskan, and C.E. Wyman, Science 344, 709 (2014). 29. J.G. Linger, D.R. Vardon, M.T. Guarnieri, E.M. Karp, G.B. REFERENCES Hunsinger, M.A. Franden, C.W. Johnson, G. Chupka, T.J. Strathmann, P.T. Pienkos, and G.T. Beckham, Proc. Natl. 1. International Reference Life Cycle Data (ILCD): ILCD Acad. Sci. U.S.A. 111, 12013 (2014). System Handbook - General Guide for Life Cycle Assess- 30. I. Klein, B. Saha, and M.M. Abu-Omar, Catal. Sci. Technol. ment - Detailed Guidance. EUR 24708 EN (Publications 5, 3242 (2015). Office of the European Union, Luxemburg, 2010), http:// 31. T. Parsell, S. Yohe, J. Degenstein, T. Jarrell, I. Klein, E. publications.jrc.ec.europa.eu/repository/bitstream/JRC48157/ Gencer, B. Hewetson, M. Hurt, J.I. Kim, H. Choudhari, B. ilcd_handbook-general_guide_for_lca-detailed_guidance_ Saha, R. Meilan, N. Mosier, F. Ribeiro, W.N. Delgass, C. 12march2010_isbn_fin.pdf. Accessed 11 April 2016. Chapple, H.I. Kentta¨maa, R. Agrawal, and M.M. Abu- 2. M.A. Curran, Curr. Opin. Chem. Eng. 2, 273 (2013). Omar, Green Chem. 17, 1492 (2015). 3. R. Bergman, M. Puettmann, A. Taylor, and K.E. Skog, For. 32. F. Derbyshire, M. Jagtoyen, and M. Thwaites, Porosity in Prod. J. 64, 220 (2014). Carbons: Characterization and Applications, ed. J.W. Pa- 4. R. Sathre and J. O’Connor, Environ. Sci. Policy 13, 104 trick (London, U.K.: Edward Arnold, 1995), p. 209. (2010). 33. M.-M. Titirici, R.J. White, N. Brun, V.L. Budarin, D.S. Su, 5. R.J. Moon, G. Schueneman and J. Simonsen, JOM, 2016. F. del Monte, J.H. Clark, and M.J. MacLachlan, Chem. Soc. doi:10.1007/s11837-016-2018-7. Rev. 44, 250 (2015). 6. A.C. Wiedenhoeft, Handbook of Wood Chemistry and Wood 34. H. Jankowska, A. S´ wia˛tkowski, and J. Choma, Active Composites, ed. R.M. Rowell (Boca Raton, FL: CRC Press, Carbon (Englewood Cliffs, NJ: Prentice-Hall, 1991). 2013), p. 9. 35. Y. Uraki and K. Koda, J. Wood Sci. 61, 447 (2015). 7. A. Panshin and C. Zeeuw, Textbook of Wood Techinology: 36. J. Spender, A.L. Demers, X. Xie, A.E. Cline, M.A. Earle, Structure, Identification, Properties, and Uses of the Com- L.D. Ellis, and D.J. Neivandt, Nano Lett. 12, 3857 (2012). mercial Woods of the United States and Canada, 4th ed. 37. D.A. Baker, N.C. Gallego, and F.S. Baker, J. Appl. Polym. (New York: McGraw-Hill, 1980). Sci. 124, 227 (2012). 8. N. Terashima, M. Yoshida, J. Hafreń, K. Fukushima, and 38. S. Iijima, Nature 354, 56 (1991). U. Westermark, Holzforschung 66, 907 (2012). 39. A.D. Maynard, P.A. Baron, M. Foley, A.A. Shvedova, E.R. 9. D.J. Cosgrove and M.C. Jarvis, Front. Plant Sci. 3, 204 Kisin, and V. Castranova, J. Toxicol. Environ. Health A 67, (2012). 87 (2004). 10. N. Terashima, K. Kitano, M. Kojima, M. Yoshida, H. Ya- 40. B. Goodell, X. Xie, Y. Qian, G. Daniel, M. Peterson, and J. mamoto, and U. Westermark, J. Wood Sci. 55, 409 (2009). Jellison, J. Nanosci. Nanotechnol. 8, 2472 (2008). 11. A.N. Fernandes, L.H. Thomas, C.M. Altaner, P. Callow, 41. S. Lehmann, Sustainability 4, 2707 (2012). V.T. Forsyth, D.C. Apperley, C.J. Kennedy, and M.C. Jar- 42. M. Green and J. Karsh, TALL WOOD—The case for tall vis, Proc. Natl. Acad. Sci. U.S.A. 108, E1195 (2011). wood buildings, Report prepared for the Canadian Wood 12. J. Hafren, T. Fujino, T. Itoh, U. Westermark, and N. Ter- Council on behalf of the Wood Enterprise Coalition and ashima, Holzforschung 54, 234 (2000). Forest Innovation Investment, Vancouver, BC, 2012. 13. L. Salmen and I. Burgert, Holzforschung 63, 121 (2009). 43. M. Mohammad, S. Gagnon, B. Douglas, and L. Podesto, 14. R.M. Rowell, R. Petterson and M.A. Tshabalala, Handbook Wood Des. Focus 22, 3 (2012). of Wood Chemistry and Wood Composites, ed. R.M. Rowell 44. N. Stark, Z. Cai and C. Carll, Wood Handbook: Wood as an (Boca Raton, FL: CRC Press, 2013), p. 33. Engineering Material, ed. R.J. Ross (USDA Forest Service, 15. S.V. Glass and S.L. Zelinka, Wood Handbook: Wood as an Forest Products Laboratory, General Technical Report Engineering Material, ed. R.J. Ross (USDA Forest Service, FPL- GTR-190, 2010), p. 11-1. Forest Products Laboratory, General Technical Report 45. F. Lam, M. Gehloff, and M. Closen, Proc. Inst. Civ. Eng FPL- GTR-190, 2010), p. 4-1. Struct. Build. 163, 267 (2010). 16. M.E. Himmel, S.-Y. Ding, D.K. Johnson, W.S. Adney, M.R. 46. A. Van Houtte, A. Buchanan, and P. Moss, Proceedings of Nimlos, J.W. Brady, and T.D. Foust, Science 315, 804 the 8th World Conference on Timber Engineering (Lahti, (2007). Finland, 2004). 17. C.E. Wyman, B.E. Dale, R.T. Elander, M. Holtzapple, M.R. 47. A. Buchanan and R. Fairweather, IPENZ Annual Confer- Ladisch, Y.Y. Lee, C. Mitchinson, and J.N. Saddler, ence 1993, Proceedings of Sustainable development Biotechnol. Prog. 25, 333 (2009). (Wellington, New Zealand, 1993) p. 343. 2404 Jakes, Arzola, Bergman, Ciesielski, Hunt, Rahbar, Tshabalala, Wiedenhoeft, and Zelinka

48. D. Mudarri and W. Fisk, Indoor Air 17, 226 (2007). 65. C. de Las Heras Alarcon, S. Pennadam, and C. Alexander, 49. V. Arantes and B. Goodell, Deterioration and Protection of Chem. Soc. Rev. 34, 276 (2005). Sustainable Biomaterials. ACS Symp. 1158, 3 (2014). 66. R. Vidoni, T. Mimmo, and C. Pandolfi, J. Bionic Eng. 12, 50. S.T. Lebow, Wood Handbook: Wood as an Engineering 250 (2015). Material, ed. R.J. Ross (USDA Forest Service, Forest 67. C. Dawson, J.F.V. Vincent, and A.-M. Rocca, Nature 390, Products Laboratory, General Technical Report FPL- GTR- 668 (1997). 190, 2010), p. 15-1. 68. I. Burgert, M. Eder, N. Gierlinger, and P. Fratzl, Planta 51. S. Lebow, USDA Forest Service, Forest Products Labora- 226, 981 (2007). tory, Research Paper FPL-RP-618 (2004). 69. N. Plaza, S.L. Zelinka, D.S. Stone, and J.E. Jakes, Smart 52. R. Ringman, A. Pilga˚rd, C. Brischke, and K. Richter, Mater. Struct. 22, 072001 (2013). Holzforschung 68, 239 (2014). 70. S.S. Kelley, T.G. Rials, and W.G. Glasser, J. Mater. Sci. 22, 53. C. Hill, Wood Modification: Chemical, Thermal and Other 617 (1987). Processes (New York: Wiley, 2006). 71. W.J. Cousins, Wood Sci. Technol. 12, 161 (1978). 54. S. Zelinka, S. Glass, and D. Stone, Wood Fiber Sci. 40, 544 72. A.-M. Olsson and L. Salmeń, The Softening Behavior of (2008). Hemicelluloses Related to Moisture (Washington, DC: 55. J.E. Jakes, N. Plaza, S.L. Zelinka, D.S. Stone, S.-C. Gleber, American Chemical Society, 2003). and S. Vogt, Proc. Of SPIE Vol. 9055, eds. A. Lakhtakia 73. K. Kulasinski, S. Keten, S. Churakov, D. Derome, and and R.J. Martin-Palma (2014), p. 90550 K. J. Crmeliet, Cellulose 21, 1103 (2014). 56. S.L. Zelinka, S.-C. Gleber, S. Vogt, G.M. Rodrı´guez Lo´pez, 74. X. Wu, R.J. Moon, and A. Martini, Cellulose 21, 2233 and J.E. Jakes, Holzforschung 69, 441 (2015). (2014). 57. J. Jakes, N. Plaza, D. Stone, and C. Hunt, J. For. Prod. Ind. 75. R. Sinko, S. Mishra, L. Ruiz, N. Brandis, and S. Keten, 2, 10 (2013). ACS Macro Lett. 3, 64 (2014). 58. E.I.S. Flores, F.A. DiazDelaO, R.M. Ajaj, M.I. Friswell, and 76. C.J. Houtman and R.H. Atalla, Plant Physiol. 107, 977 G.F. Fernando, Appl. Math. Model. 38, 3958 (2014). (1995). 59. L. Burns, A.P. Mouritz, D. Pook, and S. Feih, Compos. B 77. S. Besombes and K. Mazeau, Plant Physiol. Biochem. 43, Eng. 69, 222 (2015). 299 (2005). 60. P. Oinonen, H. Krawczyk, M. Ek, G. Henriksson, and R. 78. R.J. Keenan, G.A. Reams, F. Achard, J.V. de Freitas, A. Moriana, Carbohydr. Polym. 136, 146 (2016). Grainger, and E. Lindquist, For. Ecol. Manag. 352, 9 61. R. Merindol, S. Diabang, O. Felix, T. Roland, C. Gauthier, (2015). and G. Decher, ACS Nano 9, 1127 (2015). 79. K.G. MacDicken, P. Sola, J.E. Hall, C. Sabogal, M. Ta- 62. P.T. Mather, X.F. Luo, and I.A. Rousseau, Annu. Rev. doum, and C. de Wasseige, For. Ecol. Manag. 352, 47 Mater. Res. 39, 445 (2009). (2015). 63. C. Liu, H. Qin, and P.T. Mather, J. Mater. Chem. 17, 1543 (2007). 80. E.E. Dormontt, M. Boner, B. Braun, G. Breulmann, B. 64. M.A.C. Stuart, W.T.S. Huck, J. Genzer, M. Mu¨ller, C. Ober, Degen, E. Espinoza, S. Gardner, P. Guillery, J.C. Her- M. Stamm, G.B. Sukhorukov, I. Szleifer, V.V. Tsukruk, M. manson, G. Koch, S.L. Lee, M. Kanashiro, A. Rimbawanto, Urban, F. Winnik, S. Zauscher, I. Luzinov, and S. Minko, D. Thomas, A.C. Wiedenhoeft, Y. Yin, J. Zahnen, and A.J. Nat. Mater. 9, 101 (2010). Lowe, Biol. Conserv. 191, 790 (2015).